Voice tone quality similarity measurement method for musical instrument under tone quality migration

A similarity measurement and timbre technology, applied in instruments, speech analysis, speech recognition, etc., can solve the problem of no human voice chromatic similarity description

Inactive Publication Date: 2018-08-17
FUZHOU UNIV
View PDF11 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0015] Although the existing research on human voice timbre can effectively solve some sin...

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Voice tone quality similarity measurement method for musical instrument under tone quality migration
  • Voice tone quality similarity measurement method for musical instrument under tone quality migration
  • Voice tone quality similarity measurement method for musical instrument under tone quality migration

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] The present invention will be further explained below in conjunction with the accompanying drawings and specific embodiments.

[0031] The present invention provides a method for measuring the similarity of human voice under the transfer of musical instrument timbre, which includes the following steps: Step S1: Obtaining musical instrument audio files, constructing and training a deep convolutional neural network model, and obtaining a trained IDCNN model; Step S2 : Obtain the vocal a cappella audio file, use the human voice audio data to fine-tune the trained IDCNN model, obtain the representation model HDCNN of the human voice, and the HDCNN model outputs the timbre vector of the human voice; Step S3: Calculate the cosine degree between different timbre vectors, Get the similarity between different vocal timbres. For a schematic diagram of the main process, see figure 1 .

[0032] In an embodiment of the present invention, the step S1 specifically includes the follo...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention relates to a voice tone quality similarity measurement method for a musical instrument under tone quality migration. The method comprises the steps: firstly calculating logarithmic Mel-spectrum characteristics of sufficient and balanced musical instrument data set audio, obtaining a characteristic matrix, training a deep convolution neural network model through the musical instrumentdata, and obtaining a trained deep convolution neural network model; secondly calculating and obtaining the logarithmic Mel-spectrum characteristics of insufficient and imbalanced voice audio data set through the same method; thirdly carrying out the fine tuning of the trained deep convolution neural network model through a fine tuning method based on the transfer learning, and obtaining a voicetone representation model, wherein the output of the model is a voice tone vector; finally achieving the measurement of the similarity of voice tones through calculating the cosine similarity of tonevectors.

Description

technical field [0001] The invention belongs to the field of audio signal processing in the field of singing, in particular to a method for measuring the similarity of human voice under the transition of musical instrument timbre. Background technique [0002] The American National Institute of Standardization defines timbre as follows, "timbre refers to a certain attribute of sound produced by hearing, and the listener can judge the difference between two sounds presented in the same way with the same pitch and loudness. ". Therefore, the vocal timbre during singing refers to the voice characteristics that people use to identify the specific singer when different singers sing the same song. [0003] Spectrograms are commonly used in acoustic experiments for sound analysis. The spectrogram can display the characteristics of amplitude changing with frequency and time. The ordinate represents frequency, and the abscissa represents time. The magnitude of the amplitude is repr...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G10L25/81G10L15/06G10L15/16G10L21/0224G10L21/0232G10L21/0272G10L25/18G10L25/24G10L25/30
CPCG10L15/063G10L15/16G10L21/0224G10L21/0232G10L21/0272G10L25/18G10L25/24G10L25/30G10L25/81
Inventor 余春艳齐子铭刘煌张栋
Owner FUZHOU UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap